When a smoker confronts cues previously paired with smoking behavior, those cues trigger responses that are presumed to motivate or mediate drug use. This observation has led researchers to consider the potential therapeutic benefit of diminishing the associative impact of drug cues in humans using basic animal extinction methods. However, such cue-exposure treatments for addiction have proven less effective than anticipated. A key reason for their limited efficacy might be that human researchers have not exploited contemporary animal research that suggest new ways of thinking about extinction learning and strengthening it by guarding against specific phenomena that threaten it. Using smoking as the model drug-use behavior, the studies proposed in this R21 application for RFA #MH-04-005 will examine two such threats to extinction learning, namely renewal, which involves renewed responding to an extinguished cue when it is re-encountered in a new context after extinction training, and spontaneous recovery, which occurs when time is allowed to pass between extinction and re-exposure to a cue. The immediate objective of the proposed work is to determine whether spaced or massed presentation of cues leads to more complete within-session extinction of a salient smoking cue (Study 1). Using the optimal extinction schedule determined in Study 1, the second aim is to examine context-dependent renewal effects in human smokers by having them create personal pictorial smoking-contexts to bring into the lab. Renewal studies will determine if, as has been demonstrated in animal studies, increasing the number of contexts present during extinction (Study 2) or pairing an extinction reminder with extinction training and re-test (Study 3A) attenuates renewal effects. A final study will examine if an extinction reminder can also attenuate spontaneous recovery (Study 3B). The long-term goal of this line of research will be to create more efficacious techniques for extinguishing conditioned responding to salient drug cues in humans for the purpose of enhancing extinction-based treatments for drug addiction. Moreover, the experiments proposed here will inform pre-clinical research on nicotine self-administration in animals with regard to approaching drug-extinction in ways that more closely reflect the problems encountered in human cue-extinction research.